I don't think it's mostly about dates and times in the dataset names (at least my original post wasn't), but dates and times should be a separate discussion.
Since date and time of dataset creation are not system-dependent, the OS could supply variables and code to substitute for values in the dataset name at the time it's allocated, My suggestions for that would involve 7 variables &SYSUDTE - replaced by Dyymmdd where the date is based on UTC &SYSJDTE - replaced by Dyyyyddd where the date is Juian based on UTC &SYSUTME - replaced by Thhmmss where the time is based on UTC (and forget about it if you're creating more than one dataset of this time per second!) Three more prefixed with &SYSL where they're based on the local time. &SYSTZID - replaced by TZhhmmE or TZhhmmW to indicate the hours east or west of UTC. If you want to get a little more unwieldy, substitute with someting like Yyyyy.Mmm.Ddd for the date variable. As for the system-related static symbols, a lurker send me an elegant solution. They run, at IPL, and periodically, I think daily, a Rexx exec which retrieves the static symbols of interest and creates a member in a proc library containing SET statements for variables of other names, with the contents of the static symbols (for example, perhaps a variable named CLONESYS to contain the value of SYSCLONE). Since I'm submitting/converting/executing all on the same system, it'll work for me. Tim Hare Senior Systems Programmer Florida Department of Transportation (850) 414-4209 ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
